Quantitative assessment of eye movements using a binocular paradigm: comparison among amblyopic, cured amblyopic and normal children

doi:10.21203/rs.3.rs-1542608/v1

Download PDF

Research Article

Quantitative assessment of eye movements using a binocular paradigm: comparison among amblyopic, cured amblyopic and normal children

https://doi.org/10.21203/rs.3.rs-1542608/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: To investigate the fixation and saccade functions in children with amblyopia and cured amblyopia by a binocular eye-tracking paradigm.

Methods: Eye movements of 135 pediatric subjects, including 45 amblyopic children, 45 cured amblyopic children and 45 age-similar normal controls, were recorded under binocular viewing. The deviation (degree) between fixation positions and the visual targets along horizontal or vertical axis was recorded as the mean from both eyes. Deviations were compared across three groups and between each two groups.

Results: Main outcome measures included deviations along horizontal and vertical axes in the circular fixation test (Fix-X, Fix-Y) and horizontal saccade test (Sac-X, Sac-Y). All four deviations were significantly larger in the amblyopia group compared to the other two groups, indicating fixation and saccade deficits. However, there was no significant difference in deviations between cured amblyopic children and normal controls. In the amblyopia group, four deviations were similar across subgroups of varying amblyopia severity. Larger deviations were related to younger age and poor visual acuity overall.

Conclusion: Amblyopic children have poor fixation and saccade functions, while cured amblyopic children seem to restore the functions. This paradigm provides a promising technique to evaluate and follow up the functional recovery.

eye movement

fixation

saccade

amblyopia

binocular viewing

Fixations and saccades are two important functions of voluntary eye movements for visual perception of objects and scenes in humans. Fixation is an important function to hold the central visual field on a target for a certain period of time, which is critical to obtain sufficient details of visual targets¹. Saccades are fast eye movements to reorient gaze, changing the image of an interested object from one to another onto the fovea, which are important for everyday activities such as improving the accuracy and precision of hand reaching² and manual manipulation³, and the efficiency of reading⁴. Fixations and saccades provide humans with continuous information from the outside world.

Amblyopia is diminished vision due to inadequate visual experience during infancy and early childhood without structural eye abnormalities⁵. Abnormal eye movements in patients with amblyopia have been reported, such as increased frequency and inaccuracy of saccades in reading⁴, increased fixation instability^4,6−9, and prolonged latency and decreased precision in saccades¹⁰. The paradigms used in previous studies above were typically conducted in monocular status, while the fixation and saccade deficits of amblyopia patients under binocular viewing, similar to daily life, have been rarely reported. Only a handful of studies evaluated eye movements under binocular viewing, but analyzed the amblyopic eye or the fellow eye seperately^4,9−10 and reported monocular outcomes rather than integrated data of both eyes. However, binocular integrated data, the mean of fixation deviation from both eyes in this study, could better simulate how binocular summation¹¹ effects on eye movement performance in real world. In addition, it remains unrevealed whether timely intervention of amblyopia in childhood could restore eye movement functions while resolving visual acuity.

In this study, we described a binocular eye-tracking paradigm to assess fixation and saccade functions in three groups of participants, including amblyopic, cured amblyopic and normal children. The aims were to investigate whether eye movement functions would be impaired in amblyopia, and whether timely and effective intervention for amblyopia in childhood would simultaneously improve the functions.

Participants

This study included 135 pediatric subjects aged 4–14 years who presented to the Department of Ophthalmology at Eye & ENT Hospital of Fudan University from Jan 2021 to Jan 2022. Subjects were assigned into 3 groups according to their clinical diagnosis, including the amblyopia group (n = 45), cured amblyopia group (n = 45), and control group (n = 45). Inclusion criteria included well coordination, with or without anisometropia (the presence of interocular difference ≥ 1.00 diopters [D] in spherical equivalent [SE] or ≥ 1.50D in astigmatism in any meridian) and without strabismus (≤ 5PD at near and distance). Exclusion criteria included nystagmus, congenital or acquired ophthalmic conditions (e.g., optic nerve disease, glaucoma or cataract), systemic diseases (e.g., developmental delays, neurologic disorders), premature birth, and pool coordination. Chinese was the primary language for all children.

Amblyopia was defined as an interocular difference in best-corrected visual acuity (BCVA) of two or more lines (≥ 0.2 logMAR), or binocular BCVA worse than the lower limit considering age (0.4 logMAR for age 4 years; 0.3 logMAR for age 5 years; 0.7 logMAR for age > 5 years). Amblyopia was further classified by BCVA in the amblyopic eye as mild (< 0.3 logMAR), moderate (0.3–0.69 logMAR), and severe (≥ 0.7 logMAR) amblyopia. Patients diagnosed as initial or uncured amblyopia at the most recent visit were assigned to the amblyopia group, while patients had history of amblyopia but resolved visual acuity were assigned to the cured amblyopia group. A group of age-similar controls were also enrolled, who had normal or corrected-to-normal visual acuity in both eyes, normal stereoacuity (≥ 60" of arc), no severe ametropia (myopia worse than − 6.00D, hyperopia worse than + 5.00D, or astigmatism in any meridian ≥ 2.50D), and no history of amblyopia or strabismus.

This study observed the tenets of the Declaration of Helsinki and was approved by the institutional review board of the Eye & ENT Hospital of Fudan University. Written informed consent was obtained from the parents/guardians of the participants and assent was obtained from children ≥ 7 years of age prior to their participation.

Procedures

All the subjects received the same and detailed ophthalmic examinations in our department, including visual acuity testing, cycloplegic refraction, cornea’s optical reflection (Hirschberg test), anterior segment examination with the slit lamp, fundoscopy, ocular motility, and stereopsis. Eye movements under binocular viewing, in the circular fixation and horizontal saccade tests, were quantitatively evaluated (before cycloplegic refraction) as described below.

Apparatus and eye movement recording

Testing took place in a quiet and private room with a natural and constant luminance. A 32-inch 3-dimension (3D) monitor (resolution 1920×1080 pixels at a refresh frequency of 120Hz; LG Electronics, Seoul, Korea) was used to present stimuli at a viewing distance of 80cm. Participants were asked to wear 3D polarized glasses with their habitual optical correction (if any) and seated on a non-wheeled but height-adjustable chair with the eyes at the same level as the screen center, to make sure that the eye tracker is able to track the eyes with optimal performance. Fixation positions were recorded with a 120Hz remote eye tracker (Tobii Eye Tracker 5). The presentation of stimuli was generated by MATLAB (MathWorks, Natick, MA).

Calibration

The participant was briefly familiarized with the procedures by the experimenter, and were asked to adjust and maintain the head position until the eye tracker could catch his/her both eyes optimally. A 3-point (X, Y = 0°, + 13.5°; -13.5°, -13.5°; +13.5°, -13.5°; presented for 4 seconds at each location) calibration and validation of the eye tracker was run at the beginning of the main experiment and whenever necessary during the experiment. The calibration stimulus, a bright blue dot on a black background, was dynamically shrinking from a normal size (diameter 0.3°) and the binocular data was collected at the moment the dot disappeared. The fixation position was recorded as separate horizontal and vertical components by the eye tracker, simultaneously and respectively for both eyes. Horizontal and vertical calibration values of each eye were applied to the fixation position data. Blinks or partial blinks were detected and removed from sampling, where the pupil information was missing or there was a sudden change in pupil size. The following main tests would be initiated when calibration and its subsequent validation were acceptable with adequate accuracy for each eye.

Circular fixation test

The circular fixation test measured the deviation between fixation position and the visual target under static binocular-viewing condition to assess fixation function. The participant was instructed to fixate a visual target on a black background, which was a bright blue dot (1.4° diameter) with a black cross-shaped center. The visual target appeared sequentially at 9 locations, 8 of which were located on the outer circle (8.3° radius) and one was located at the center of the circle (Fig. 1a). It presented for 3 seconds at each location and automatically switched to the next one, while data recording was manually initiated by the experimenter at each location after accurate and stable fixation of the participant. At each location, the eye tracker extracted 5 samples from all real-time fixation positions it grabbed, calculated the average deviation (separately along horizontal and vertical axes) between fixation positions and the visual target for each eye individually, and recorded as the binocular mean values (Fig. 2a).

Horizontal saccade test

The horizontal saccade test measured the deviation between post-saccadic fixation positions and the visual target under dynamic binocular-viewing condition to assess saccade function. A bright blue cross-shaped target (1.4° diameter) appeared at 8 locations on a black background in turn, ± 17.8° horizontally alternately and + 8.3°, + 2.8°, -2.8°, -8.3° vertically sequentially, requiring a wide-range ocular motion (saccade) and succeeding fixation (Fig. 1b). The participant was instructed to saccade and then fixate the target as quickly and accurately as possible. The visual target presented for 3 seconds at each location and switched to the next one with data recording initiated automatically. Considering saccade latency and saccade initiation¹⁰ during the target switching, the eye tracker only grabbed real-time fixation positions from 800ms to 1800ms at each location, and extracted 10 from all samples. Likewise, at each location, average deviations (separately along horizontal and vertical axes) were calculated for each eye individually and recorded as the binocular mean values (Fig. 2b).

Data analysis

The horizontal and vertical deviations between fixation positions and the visual target in the two tests were calculated for every participant by equations below.

Horizontal deviation =\(\frac{1}{k}\sum _{1\le j\le k} \frac{1}{\text{n}}\sum _{1\le i\le n}\frac{\left({\Delta }{\text{X}}_{Left}+ {\Delta }{\text{X}}_{Right}\right)}{2}\)

Vertical deviation =\(\frac{1}{k}\sum _{1\le j\le k} \frac{1}{\text{n}}\sum _{1\le i\le n}\frac{\left({\Delta }{\text{Y}}_{Left}+ {\Delta }{\text{Y}}_{Right}\right)}{2}\)

In the equations, the constant k was the number of locations where the visual target would appear in the test; and the constant n was the total number of fixation positions extracted to calculate average deviation for each location. As test procedures described above, constants k and n were 9 and 5 in the circular fixation test, and 8 and 10 in the horizontal saccade test. ΔX and ΔY represented the horizontal and vertical deviation (°) between each fixation position extracted and the visual target, with subscripts Left and Right representing the left and right eye.

For simplicity, horizontal and vertical deviations (°) were abbreviated as Fix-X and Fix-Y in the circular fixation test, and Sac-X and Sac-Y in the horizontal saccade test, respectively. The smaller the deviations measured, the more accurate and controllable the fixation or saccade function was.

Normally distributed continuous data were expressed as means ± standard deviations (SD). Abnormally distributed continuous data were expressed as median (25th percentile [P25], 75th percentile [P75]). Categorical variables were described as frequency counts and proportions (%). Differences in baseline characteristics among multiple groups were evaluated by one-way analysis of variance (ANOVA) for continuous variables with normal distribution, nonparametric Kruskal-Wallis test for abnormal distribution, and Pearson’s χ² test or Fisher’s exact test for categorical variables.

Since deviations were distributed right-skewed, Kruskal-Wallis tests were performed to compare the four deviations across three independent groups. Pairwise comparisons (Kruskal-Wallis ANOVA) were conducted to determine differences between each two groups, and significance values were Bonferroni corrected for multiple tests. Kruskal-Wallis tests were also used to compare the four deviations across subgroups of varying amblyopia severity (mild, moderate and severe amblyopia). Mann-Whitney U tests were used to compare deviations in control subjects with and without anisometropia. Spearman’s rank correlations were performed to evaluate the correlation between clinical characteristics (age, BCVA in amblyopic/fellow eye, anisometropia) and the four deviations. Wilcoxon signed-rank tests were used to evaluate repeatability and stability of the methodology. The significance level was set at p < 0.05 (two-sided probability). All statistical analyses were performed using SPSS version 26 (SPSS Inc, Chicago, IL, USA), and graphs were constructed by GraphPad Prism 8 (GraphPad Software Inc., San Diego CA).

Clinical and demographic characteristics of the three groups were showed in Table 1. The mean age of total participants was 7.21 ± 2.40 years (range, 4–14 years). Three groups were similar in age (p = 0.172) and sex (p = 0.473).

Table 1

Clinical and demographic characteristics in three groups.
	Amblyopia	Cured Amblyopia	Control	p value
	(n=45)	(n=45)	(n=45)	p value
Age (years)	6.67±2.31	7.56±2.45	7.40±2.40	0.172^d
Sex: Female (%)	23(51.1)	23(51.1)	28(62.2)	0.473^c
Amblyopia Severity (%)				N/A
No amblyopia	N/A	45(100)	45(100)
mild amblyopia	23(51.1)	N/A	N/A
moderate amblyopia	13(28.9)	N/A	N/A
severe amblyopia	9(20.0)	N/A	N/A
Amblyopia type (%)				N/A
Unilateral	40(88.9)	N/A	N/A
Bilateral	5(11.1)	N/A	N/A
BCVA in the amblyopic/right eye (logMAR)^a	0.37±0.30	0.04±0.06	0.02±0.04	<0.001^e
BCVA in the fellow/left eye (LogMAR)^b	0.08±0.12	0.03±0.06	0.02±0.04	0.007^e
SE in the amblyopic/right eye (D)^a	+2.64±3.63	+2.20±2.71	-0.50±1.97	<0.001^e
SE in the fellow/left eye (D)^b	+1.19±2.64	+1.46±2.04	-0.29±1.90	<0.001^e
SE of refraction (D)^c				<0.001^f
<-6.00	2(4.4)	0(0)	0(0)
-6.00~-3.00	1(2.2)	2(4.4)	4(8.9)
-3.00~-0.50	5(11.1)	4(8.9)	17(37.8)
-0.50~+0.50	0(0)	3(6.7)	11(24.4)
+0.50~+3.00	13(28.9)	20(44.4)	9(20.0)
+3.00~+6.00	21(46.7)	13(28.9)	4(28.9)
>+6.00	3(6.7)	3(6.7)	0(0)
Anisometropia (%)	26(57.8)	20(44.4)	10(22.2)	0.003^f
F, female; BCVA, best-corrected visual acuity; logMAR, logarithm of the minimum angle of resolution; SE, spherical equivalent, algebraic sum of the dioptric power of the sphere and half of the cylinder; D, diopter; N/A, not applicable. ^athe amblyopic eye in the amblyopia group, the previously amblyopic eye in the cured amblyopia group, and the right eye in the control group; ^b the fellow eye in the amblyopia group, the previously fellow eye in the cured amblyopia group, and the left eye in the control group; ^c the higher absolute value of SE in both eyes was used for refraction gradin; ^d analyzed by one-way ANOVA; ^e analyzed by Kruskal-Wallis test; ^f analyzed by the Pearson’s χ2 test or Fisher’s exact test. p<0.05 indicated significant difference among three groups.

There were significant differences in BCVA and SE in each eye among the three groups (Table 1). According to pairwise comparisons, BCVA in the amblyopic and fellow eye in the amblyopia group were larger than that in the right and left eye in the control group (K-W H = 67.833, Adj. p < 0.001; H = 19.711, Adj. p = 0.012), and also larger than that in the previously amblyopic and fellow eye in the cured amblyopic group (H = 58.400, Adj. p < 0.001; H = 17.622, Adj. p = 0.03); while the differences between the control group and the cured amblyopia group were not statistically significant (H = 9.433, Adj. p = 0.676; H = 2.089, Adj. p = 1.000). Compared to SE in the right and left eye in the control group, larger SE was showed in the amblyopic and fellow eye in the amblyopia group (H = 47.389, Adj. p < 0.001; H = 32.667, Adj. p < 0.001), and also in the previously amblyopic and fellow eye in the cured amblyopia group (H = 38.544, Adj. p < 0.001; H = 32.533, Adj. p < 0.001); while the differences between the amblyopia group and the cured amblyopia group were not statistically significant (H = 8.844, Adj. p = 0.85; H = 0.133, Adj. p = 1.000).

Main outcome measures included four deviations (°) for every participant, including Fix-X and Fix-Y in the circular fixation test, as well as Sac-X and Sac-Y in the horizontal saccade test. The four deviations were positively correlated with each other (p < 0.05), both in total and within each group. Deviation data was expressed as median (P25, P75).

Deviations in normal controls

In the control group, four deviations (median [P25, P75]) in the paradigm, Fix-X, Fix-Y, Sac-X and Sac-Y, were 1.58 (0.84, 2.60)°, 2.18 (1.56, 2.61)°, 3.28 (1.79, 4.19)° and 2.48 (1.71, 3.65)°, respectively. There was no significant difference in Fix-X (Mann-Whitney U = 125, p = 0.179), Fix-Y (U = 139.5, p = 0.338), Sac-X (U = 171, p = 0.925) and Sac-Y (U = 148.5, p = 0.475) between subjects with (n = 10) and without anisometropia (n = 35), who were also similar in age (p = 0.055) and sex (p = 0.197).

Well functions in normal controls could be directly visualized from test results on the screen, showing fixation positions of highly accuracy and precision in the two tests (Fig. 2a, Fig. 2b).

Deviations in amblyopia

In the circular fixation test, Fix-X and Fix-Y in the amblyopia group were 2.58 (1.40, 5.30)° and 3.10 (2.19, 4.50)°, respectively. Amblyopic children had significantly larger Fix-X (mean ± SEM 3.25 ± 0.32°) and Fix-Y (3.45 ± 0.21°) than normal controls (1.78 ± 0.16°, H = 27.656, Adj. p = 0.002; 2.16°±0.12°, H = 34.689, Adj. p < 0.001), indicating fixation deficit with poor accuracy in static fixations along both the horizontal and vertical axes in amblyopia.

In the horizontal saccade test, Sac-X and Sac-Y in the amblyopia group were 7.35 (4.93, 10.01)° and 4.45 (3.15, 5.85)°, respectively. Amblyopic children had larger Sac-X (mean ± SEM 7.65°±0.53°) and Sac-Y (4.64°±0.27°) than normal controls (3.19°±0.22°, H = 51.078, Adj. p < 0.001; 2.81°±0.20°, H = 40.589, Adj. p < 0.001) as well, indicating saccade deficit with poor accuracy of post-saccadic fixations along both the horizontal and vertical axes in amblyopia.

There was no significant difference in deviations across subgroups of varying amblyopia severity (K-W, p > 0.05).

Deviations in cured amblyopia

In the circular fixation test, Fix-X and Fix-Y in the cured amblyopia group were 1.45 (0.83, 2.96)° and 1.93 (1.53, 3.20)°, respectively. Fix-X and Fix-Y of cured amblyopic children (mean ± SEM 1.98 ± 0.21°; 2.42°±0.19°) were similar with normal controls (H = 2.411, Adj. p = 1.000; H = 3.511, Adj. p = 1.000), but significantly smaller than those of amblyopic children (H = 25.244, Adj. p = 0.007; H = 31.178, Adj. p < 0.001), indicating greatly improved accuracy and controllability of static fixations in cured amblyopia.

In the horizontal saccade test, Sac-X and Sac-Y in the cured amblyopia group were 3.78 (2.38, 5.93)° and 2.98 (2.11, 4.96)°, respectively. Similarly, Sac-X and Sac-Y of cured amblyopic children (mean ± SEM 4.20 ± 0.37°; 3.54 ± 0.29°) were similar with those of normal controls (H = 13.156, Adj. p = 0.332; H = 15.944, Adj. p = 0.159), but significantly smaller than those of amblyopic children (H = 37.922, Adj. p < 0.001; H = 24.644, Adj. p = 0.008), indicating greatly improved accuracy and controllability of post-saccadic fixations in cured amblyopia.

All four deviations were significant difference across the three groups (Table 2). The mean value of each deviation was found in the same decreasing order: amblyopia group > cured amblyopia group > control group. According to pairwise comparisons, the distribution of each deviation was significantly different between the amblyopia group and the other two groups, but not between the cured amblyopia group and the control group (Fig. 3).

Table 2

Deviations in the fixation and saccade tests in the three groups.
	Control	Amblyopia	Cured Amblyopia	p value
Fixation
Fix-X	2.58 (1.40, 5.30)	1.45 (0.83, 2.96)	1.58 (0.84, 2.60)	0.001^a
Fix-Y	3.10 (2.19, 4.50)	1.93 (1.53, 3.20)	2.18 (1.56, 2.61)	< 0.001^a
Saccade
Sac-X	7.35 (4.93, 10.01)	3.78 (2.38, 5.93)	3.28 (1.79, 4.19)	< 0.001 ^a
Sac-Y	4.45 (3.15, 5.85)	2.98 (2.11, 4.96)	2.48 (1.71, 3.65)	< 0.001^a
Deviation (°) was expressed as median (P25, P75). Fix-X/ Fix-Y: horizontal/ vertical deviation of the circular fixation test; Sac-X/ Sac-Y: horizontal/ vertical deviation of the horizontal saccade test.
^a analyzed by Kruskal-Wallis tests, p < 0.05 indicating significantly different distributions of deviations among three groups.

Compared with normal controls, fixation and saccade deficits in amblyopic children could also be directly visualized from the results on the screen (Fig. 2c, Fig. 2d), showing large distance between fixation positions and the visual target and large discreteness within fixation positions at each location, namely poor accuracy and precision in static and post-saccadic fixation, which seemed to be largely restored in cured amblyopic children (Fig. 2e, Fig. 2f). Since the results displayed on the screen immediately at the end of each test, a re-measurement was required once the results seemed unreliable due to the participant’s inattention during the test.

Repeated experiments, 20 minutes apart, were conducted for 102 well-coordinated participants (41 controls, 26 amblyopia, and 35 cured amblyopia), and their data was used to validate the stability of this method and rule out potential learning effects. Repeated measures for all the four deviations had similar results in all 102 participants and separate groups (Table S1).

Relationships between deviations and clinical characteristics

Larger deviations were significantly correlated to younger age (n = 135, Fix-X R_s=-0.209, p = 0.015; Fix-Y R_s=-0.215, p = 0.012; Sac-X R_s=-0.256, p = 0.003; Sac-Y R_s=-0.263, p = 0.002) (Fig. 4a). Speaking of each group, only 3 of 4 deviations (except Fix-Y) in cured amblyopic children were significantly correlated to age, while there was no correlation between age and deviations in amblyopic children and normal controls.

Larger deviations were significantly correlated to poor visual acuity, not only to BCVA in the amblyopic/right eye (n = 135, Fix-X R_s=0.335, p < 0.001; Fix-Y R_s=0.336, p < 0.001; Sac-X R_s=0.504, p < 0.001; Sac-Y R_s=0.404, p < 0.001) (Fig. 4b), but also to BCVA in the fellow/left eye (n = 135, Fix-X R_s=0.219, p = 0.011; Fix-Y R_s=0.181, p = 0.036; Sac-X R_s=0.236, p = 0.006; Sac-Y R_s=0.226, p = 0.008) (Fig. 4c). However, speaking of each group, significant correlations were only found between Fix-X and BCVA in each eye in normal controls, and between Sac-X and BCVA in the previously amblyopic eye in cured amblyopic children.

Sac-X and Sac-Y were significantly correlated to larger SE in the amblyopic/right eye (n = 135, R_s=0.180, p = 0.036; R_s=0.239, p = 0.005). There was no significant correlation between deviations and sex or anisometropia, either overall or within each group.

In this study, we assessed fixation and saccade functions in children with amblyopia and cured amblyopia, using a binocular eye-tracking diagram. Our main findings are as follows. (1) Amblyopic children had eye movement deficits compared with cured amblyopic children and normal subjects. (2) There was no significant difference in fixation and saccade functions between cured amblyopic children and normal subjects. (3) Overall, larger deviations were correlated to younger age and poor visual acuity.

Amblyopia is traditionally considered as a monocular disease, but has been increasingly appreciated as a binocular disorder over the past few decades, which results from decorrelation of binocular stimulation or discordant visual experience during visual development⁵. Children with amblyopia have not only impaired monocular visual acuity, but also affected binocularity¹², eye-hand coordination³, as well as eye movements both in the amblyopic eye and fellow eye¹³. Eye movement abnormalities in amblyopia have been reported in previous studies, but were typically assessed for each eye individually, under monocular viewing by the amblyopic eye or the fellow eye^4,6−10. However, we should focus on binocular skill deficits in amblyopia, rather than in monocular status, since eye movement tasks are habitually performed with both eyes open in daily life. Several studies were conducted under binocular viewing condition, but evaluated each eye separately^4,9−10. In the present study, we assessed deviations of static and post-saccadic fixations under binocular viewing, and significantly larger binocular mean deviations were found in amblyopic children compared to cured amblyopic children and normal controls, indicating fixation and saccade deficits with increased positional inaccuracy and poor controllability in children with amblyopia.

In the clinic, amblyopia treatment is commonly terminated once normal visual acuity achieved, though binocular deficits persist in almost half of those with recovered visual acuity, including binocular perception eye position, interocular suppression, and stereoacuity^12,14−17. However, few studies evaluated recovery of eye movement functions of those patients. Therefore, in addition to children with amblyopia, our study also included cured amblyopic children to explore whether eye movement deficits would be restored with resolved visual acuity after early intervention of amblyopia in childhood. In the present study, we found similar fixation deviations between cured amblyopic children and normal controls, indicating that early and efficient intervention may additionally restore impaired eye movements in amblyopic children.

A typical type of eye movement deficits was increased fixation instability, which has been reported previously to be associated with severity of amblyopia^7,18. One hypothesis is that, in addition to the effect of decorrelated visual experience, the monocular acuity deficit has a direct effect on fixation instability⁷. However, some previous studies also found no correlation between fixation instability and amblyopia severity¹⁹, while binocular visuomotor performance of amblyopic subjects tended to deteriorate with their loss of depth vision (i.e., stereoacuity)¹¹. In our study, lager deviations were correlated to poorer visual acuity overall, whereas not in separate groups. In children with amblyopia, the four deviations did not differ across subgroups of varying amblyopia severity. The absence of correlation between deviations and amblyopia depth might be explained by following reasons. In this study, fixation deviations were recorded as binocular mean values rather than monocular data, and hence binocular summation¹¹ could have a potential effect on binocular performance. Besides, different types of visual stimulus may have a different effect on assessment results^20–22. In addition, the sample size in each subgroup of amblyopia severity was limited and unbalanced, which could be improved in further studies with more patients included.

Lager deviations were correlated to younger age in total and the cured amblyopia group. Possible reasons for this included older children had binocular function of more complete development²³ and they were more able to concentrate on the experiment and cooperate better.

Based on our findings, timely intervention might reverse eye movement deficits simultaneously with resolving visual acuity. Standard-of-care treatments for amblyopia include monocular patching, atropine penalization and Bangerter filters²⁴. Recent evidence suggests that interocular suppression plays a major role in the etiology of amblyopia^24–26, which can be reduced by rebalancing the signals between the two eyes, thereby allowing for binocular interaction^25,27. Therefore, practical tools are also needed in clinic to assess and follow up binocular performance. Compared with previous techniques such as a microperimeter or scanning laser ophthalmoscope^7,23,28−29 or a high-resolution video-based eye tracker^5,10, strengths of this binocular eye-tracking paradigm include interactivity and comfort to facilitate effective cooperation in pediatric patients; objective, understandable and visualized results; and relatively low requirement to operate the system, making it more practicable and efficient in clinic. This paradigm provides a promising technique to evaluate and follow up the binocular functional recovery. Primary shortcoming of this paradigm is limited applicable people, for instance, patients with marked strabismus were excluded in this study because the difference between the fixation coordinates provided by their left and right eye might be quite large due to severe misalignment of the visual axes, thus the eye tracker could even record monocular data only. However, this paradigm was also applicable to those patients with strabismus realigned by surgical correction, to assess functional prognosis under binocular viewing postoperatively. Based on our study, further multi-center studies with larger sample size will help to confirm the reliability and practicability of this method in eye movement assessment for pediatric patients, and further summarize the reference range of test results in vision-healthy and amblyopic children.

In conclusion, amblyopic children have poor fixation and saccade functions, while cured amblyopic children seemed to restore the functions. This binocular eye-tracking paradigm provides a promising technique to evaluate and follow up the functional recovery.

Acknowledgments

We would like to thank very much to the children and their parents/guardians who kindly consented to participate in this study.

Authors’ contributions

C.Z. and W.W. conceived and designed the experiments. Y.Z. and H.B. performed the experiments. Y.Z., H.B. and X.Y. analyzed the data. X.Y. and H.B. contributed reagents/materials/analysis tools. C.Z., W.W. and Y.Z. discussed the results and critically reviewed of the manuscript. Y.Z. wrote the main manuscript text and prepared figures. All authors reviewed the manuscript.

Funding

Supported by National Natural Science Foundation of China (Grant No. 82020108006 and 81730025) to C.Z., National Natural Science Foundation of China (Grant No. 81500752) and Shanghai Science and Technology Innovation Action Plan (Grant No. 22ZR1410200) to W.W..

Availability of data and materials

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Ethics approval and consent to participant

This study followed the Declaration of Helsinki and was approved by the ethics committee of the Eye & ENT Hospital of Fudan University. Written informed consent was obtained from the parents/guardians of the participants included (all under the age of 16 years) prior to their participation.

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

BARLOW HB. Eye movements during fixation. J Physiol. 1952;116(3):290–306. https://doi.org/10.1113/jphysiol.1952.sp004706.
Niechwiej-Szwedo E, Goltz HC, Colpa L, Chandrakumar M, Wong AM. Effects of Reduced Acuity and Stereo Acuity on Saccades and Reaching Movements in Adults With Amblyopia and Strabismus. Invest Ophthalmol Vis Sci. 2017;58(2):914–21. https://doi.org/10.1167/iovs.16-20727.
Johansson RS, Westling G, Backstrom A, Flanagan JR. Eye-hand coordination in object manipulation. J Neurosci. 2001;21(17):6917–32.
Kelly KR, Jost RM, De La Cruz A, Dao L, Beauchamp CL, Stager DJ, et al. Slow reading in children with anisometropic amblyopia is associated with fixation instability and increased saccades. J AAPOS 2017, 21(6):447–451. https://doi.org/10.1016/j.jaapos.2017.10.001.
McKee SP, Levi DM, Movshon JA. The pattern of visual deficits in amblyopia. J Vis. 2003;3(5):380–405.
Birch EE, Subramanian V, Weakley DR. Fixation instability in anisometropic children with reduced stereopsis. J AAPOS. 2013;17(3):287–90. https://doi.org/10.1016/j.jaapos.2013.03.011.
Subramanian V, Jost RM, Birch EE. A quantitative study of fixation stability in amblyopia. Invest Ophthalmol Vis Sci. 2013;54(3):1998–2003. https://doi.org/10.1167/iovs.12-11054.
Shaikh AG, Otero-Millan J, Kumar P, Ghasia FF. Abnormal Fixational Eye Movements in Amblyopia. PLoS One. 2016;11(3):e149953. https://doi.org/10.1371/journal.pone.0149953.
Kelly KR, Cheng-Patel CS, Jost RM, Wang YZ, Birch EE. Fixation instability during binocular viewing in anisometropic and strabismic children. Exp Eye Res. 2019;183:29–37. https://doi.org/10.1016/j.exer.2018.07.013.
Niechwiej-Szwedo E, Goltz HC, Chandrakumar M, Hirji ZA, Wong AM. Effects of anisometropic amblyopia on visuomotor behavior, I: saccadic eye movements. Invest Ophthalmol Vis Sci. 2010;51(12):6348–54. https://doi.org/10.1167/iovs.10-5882.
Grant S, Moseley MJ. Amblyopia and real-world visuomotor tasks. Strabismus. 2011;19(3):119–28. https://doi.org/10.3109/09273972.2011.600423.
Birch EE. Amblyopia and binocular vision. Prog Retin Eye Res. 2013;33:67–84. https://doi.org/10.1016/j.preteyeres.2012.11.001.
Ghasia FF, Otero-Millan J, Shaikh AG. Abnormal fixational eye movements in strabismus. Br J Ophthalmol. 2018;102(2):253–9. https://doi.org/10.1136/bjophthalmol-2017-310346.
Melmoth DR, Finlay AL, Morgan MJ, Grant S. Grasping deficits and adaptations in adults with stereo vision losses. Invest Ophthalmol Vis Sci. 2009;50(8):3711–20. https://doi.org/10.1167/iovs.08-3229.
Cheng Juanjuan, Fu Jing, Lu Wei, Chu Hang, Yan Li. [The binocular visual perception function after cure of amblyopia in children]. Ophthalmol CHN. 2017;26(5):299–302. https://doi.org/10.13281/j.cnki.issn.1004-4469.2017.05.003
Wallace DK, Lazar EL, Melia M, Birch EE, Holmes JM, Hopkins KB, et al. Stereoacuity in children with anisometropic amblyopia. J AAPOS. 2011;15(5):455–61. https://doi.org/10.1016/j.jaapos.2011.06.007.
Birch EE, Wang J. Stereoacuity outcomes after treatment of infantile and accommodative esotropia. Optom Vis Sci. 2009;86(6):647–52. https://doi.org/10.1097/OPX.0b013e3181a6168d.
Carpineto P, Ciancaglini M, Nubile M, Di Marzio G, Toto L, Di Antonio L, et al. Fixation patterns evaluation by means of MP-1 microperimeter in microstrabismic children treated for unilateral amblyopia. Eur J Ophthalmol. 2007;17(6):885–90. https://doi.org/10.1177/112067210701700603.
Gonzalez EG, Wong AM, Niechwiej-Szwedo E, Tarita-Nistor L, Steinbach MJ. Eye position stability in amblyopia and in normal binocular vision. Invest Ophthalmol Vis Sci. 2012;53(9):5386–94. https://doi.org/10.1167/iovs.12-9941.
Thaler L, Schutz AC, Goodale MA, Gegenfurtner KR. What is the best fixation target? The effect of target shape on stability of fixational eye movements. Vision Res. 2013;76:31–42. https://doi.org/10.1016/j.visres.2012.10.012.
Steinman RM. Effect of target size, luminance, and color on monocular fixation. Journal of the Optical Society of America. 1965;55:1158–65. https://doi.org/10.1364/JOSA.55.001158.
McSorley E, Findlay JM. Saccade target selection in visual search: accuracy improves when more distractors are present. J Vis. 2003;3(11):877–92. https://doi.org/10.1167/3.11.20.
Tychsen L, Richards M, Wong A, Foeller P, Bradley D, Burkhalter A. The neural mechanism for Latent (fusion maldevelopment) nystagmus. J Neuroophthalmol. 2010;30(3):276–83. https://doi.org/10.1097/WNO.0b013e3181dfa9ca.
Birch EE, Kelly KR, Wang J. Recent Advances in Screening and Treatment for Amblyopia. Ophthalmol Ther. 2021;10(4):815 – 30. https://doi.org/10.1007/s40123-021-00394-7.
Hess RF, Thompson B. Amblyopia and the binocular approach to its therapy. Vision Res. 2015;114:4–16. https://doi.org/10.1016/j.visres.2015.02.009.
Mansouri B, Thompson B, Hess RF. Measurement of suprathreshold binocular interactions in amblyopia. Vision Res. 2008;48(28):2775–84. https://doi.org/10.1016/j.visres.2008.09.002.
Levi DM. Rethinking amblyopia 2020. Vision Res. 2020;176:118–29. https://doi.org/10.1016/j.visres.2020.07.014.
Crossland MD, Sims M, Galbraith RF, Rubin GS. Evaluation of a new quantitative technique to assess the number and extent of preferred retinal loci in macular disease. Vision Res. 2004;44(13):1537–46. https://doi.org/10.1016/j.visres.2004.01.006.
Molina-Martin A, Pinero DP, Perez-Cambrodi RJ. Reliability and Intersession Agreement of Microperimetric and Fixation Measurements Obtained with a New Microperimeter in Normal Eyes. Curr Eye Res. 2016;41(3):400–9. https://doi.org/10.3109/02713683.2015.1020170.

No competing interests reported.

TableS1.xlsx

Download PDF

Editorial decision: Major revision
14 Jun, 2022
Reviewers agreed at journal
30 May, 2022
Reviews received at journal
26 Apr, 2022
Reviewers agreed at journal
15 Apr, 2022
Reviewers invited by journal
11 Apr, 2022
Editor assigned by journal
11 Apr, 2022
Editor invited by journal
11 Apr, 2022
Submission checks completed at journal
11 Apr, 2022
First submitted to journal
10 Apr, 2022

You are reading this latest preprint version

Quantitative assessment of eye movements using a binocular paradigm: comparison among amblyopic, cured amblyopic and normal children

Status:

Version 1

Abstract

Figures

Introduction

Methods

Participants

Procedures

Apparatus and eye movement recording

Calibration

Circular fixation test

Horizontal saccade test

Data analysis

Results

Deviations in normal controls

Deviations in amblyopia

Deviations in cured amblyopia

Relationships between deviations and clinical characteristics

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1